Effectiveness of Kd- Tree in Ray Tracing of Dynamic Point Clouds

A kd-tree is basically defined as a data structure which is mainly used for accumulating a fixed set of points in a k-dimensional space. Kd-tree is predominantly a binary tree. Kd-tree is attracting an increasing interest in recent decades for ray tracing due to its advantages in generating detailed images based on natural physical behaviors, flexibility, and ease of coding. Both point-based representations and ray tracing provide means to efficiently show very complicated 3D models. Computational effectiveness has been the main focus of past work on ray tracing point-sampled surfaces. Kd-tree solves the complication in searching nearest neighbors of a given point in k dimensional space. Kd-trees are also used in pattern recognition and machine learning. Nearest neighbor search algorithm on kd-trees can be slightly improved to allow incremental nearest neighbor search. Kd-tree was invented by john Louis Bentley and it is an abstraction of a binary search tree. It is mainly used in a large amount of graphics applications, inclusive of nearest neighbor search in point cloud (Set of data point) modeling. Kd-tree is also used for accumulating a group of points in the Cartesian co-ordinate plane means in 3-D space. Kd-tree establishment can also be cast-off for dynamic point clouds for accelerating the nearest neighbor queries. Kd-trees are mainly used for Nearest neighbor searching, Query processing in sensor network, Database searching using multiple keys, Fingerprint matching and Optimization ray tracing.